Psychosocial Experiences of Spanish-Speaking Parents of Children With Craniofacial Microsomia.

Craniofacial microsomia (CFM) and microtia psychosocial research in the US is primarily with English-speaking participants. Given that 19% of the US is Latino, and there is a higher prevalence of CFM in Latino populations, this study aims to describe psychosocial experiences related to CFM among Spanish-speaking Latino caregivers to better inform health care. Narrative interviews (mean 73±17 min) were completed in Spanish with parents of children with CFM aged 3 to 17 (mean age 10.8±4.8 years). Transcripts were analyzed using quantitative linguistic analyses and reflexive thematic analysis. Participants (N=12) were mostly mothers (83%) who had immigrated to the US and had low socioeconomic status. Based upon analysis of grouped word counts, participants spent approximately half of their narratives discussing the first two years of their child's life. Themes selected based on US Latino sociodemographics and cultural values included the Impact of Language, Healthcare Challenges, Supportive Healthcare Experiences, Caregiver Coping with CFM, Family Roles, and Addressing Social Implications of CFM. Results highlighted that the first years of care are of critical importance to parents and suggest this is an optimal time to focus on education and support services for families. Additional treatment suggestions include providing interpretation and informational materials in Spanish, addressing care barriers, supporting familial and child coping, accounting for the role of extended family, and helping address social concerns. Ongoing research with Latino families can further assist in guiding culturally sensitive CFM health care.

Effects of dietary palm oil supplementation on ruminal degradation and apparent digestibility of nutrients in sheep.

The aim of this study was to evaluate the effects of the inclusion of palm oil on the ruminal environment and nutrient digestibility of sheep diets. Twenty rumen-cannulated sheep were kept in individual stalls equipped with feeding and drinking troughs The animals were fed five diets based on Elephant grass (Pennisetum purpureum Schum. cv. Roxo) silage and supplemented with 0, 25, 50, 75, or 100 g kg-1 of palm oil (based on total DM). The Elephant grass was harvested at 90 days of regrowth and the concentrate was based on ground corn grain, soybean meal and mineral mix (20 g kg-1 DM), offered to the sheep at a ratio of 1.5 g kg-1d-1 of body weight (restricted intake) to maintain a forage-to-concentrate ratio of 1:1, based on DM. There were no differences (P = 0.324) in ruminal disappearance and degradability parameters with up to 75 g of oil per kg of DM. Organic matter showed a linear reduction in apparent digestibility, while ether extract increased linearly. Palm oil affected the digestibility and nutritional parameters in ruminant diets.

Revolutionising food advertising monitoring: a machine learning-based method for automated classification of food videos.

OBJECTIVE: Food advertising is an important determinant of unhealthy eating. However, analysing a large number of advertisements (ads) to distinguish between food and non-food content is a challenging task. This study aims to develop a machine learning-based method to automatically identify and classify food and non-food ad videos. DESIGN: Methodological study to develop an algorithm model that prioritises both accuracy and efficiency in monitoring and classifying advertising videos. SETTING: From a collection of Brazilian television (TV) ads data, we created a database and split it into three sub-databases (i.e. training, validation and test) by extracting frames from ads. Subsequently, the training database was classified using the EfficientNet neural network. The best models and data-balancing strategies were investigated using the validation database. Finally, the test database was used to apply the best model and strategy, and results were verified with field experts. PARTICIPANTS: The study used 2124 recorded Brazilian TV programming hours from 2018 to 2020. It included 703 food ads and over 20 000 non-food ads, following the protocol developed by the INFORMAS network for monitoring food marketing on TV. RESULTS: The results showed that the EfficientNet neural network associated with the balanced batches strategy achieved an overall accuracy of 90·5 % on the test database, which represents a reduction of 99·9 % of the time spent on identifying and classifying ads. CONCLUSIONS: The method studied represents a promising approach for differentiating food and non-food-related video within monitoring food marketing, which has significant practical implications for researchers, public health policymakers, and regulatory bodies.

A survey on deep learning for skin lesion segmentation.

Skin cancer is a major public health problem that could benefit from computer-aided diagnosis to reduce the burden of this common disease. Skin lesion segmentation from images is an important step toward achieving this goal. However, the presence of natural and artificial artifacts (e.g., hair and air bubbles), intrinsic factors (e.g., lesion shape and contrast), and variations in image acquisition conditions make skin lesion segmentation a challenging task. Recently, various researchers have explored the applicability of deep learning models to skin lesion segmentation. In this survey, we cross-examine 177 research papers that deal with deep learning-based segmentation of skin lesions. We analyze these works along several dimensions, including input data (datasets, preprocessing, and synthetic data generation), model design (architecture, modules, and losses), and evaluation aspects (data annotation requirements and segmentation performance). We discuss these dimensions both from the viewpoint of select seminal works, and from a systematic viewpoint, examining how those choices have influenced current trends, and how their limitations should be addressed. To facilitate comparisons, we summarize all examined works in a comprehensive table as well as an interactive table available online3.

Tumor antigen-unbiased variable epitope library contains mimotopes with antitumor effect in a mouse model of breast cancer.

Breast cancer is one of the leading causes of death that affects the female population worldwide. Despite advances in treatments and a greater understanding of the disease, there are still difficulties in successfully treating patients. Currently, the main challenge in the field of cancer vaccines is antigenic variability which can reduce antigen-specific T- cell response efficacy. The search for and validation of immunogenic antigen targets increased dramatically over the past few decades and, with the advent of modern sequencing techniques, permitting the fast and accurate identification of the neoantigen landscape of tumor cells, will undoubtedly continue to grow exponentially for years to come. We have previously implemented Variable Epitope Libraries (VEL) as an unconventional vaccine strategy in preclinical models and for identifying and selecting mutant epitope variants. Here, we used an alanine-based sequence to generate a 9-mer VEL-like combinatorial mimotope library G3d as a new class of vaccine immunogen. An in silico analysis of the 16,000 G3d-derived sequences revealed potential MHC-I binders and immunogenic mimotopes. We demonstrated the antitumor effect of treatment with G3d in the 4T1 murine model of breast cancer. Moreover, two different T cell proliferation screening assays against a panel of randomly selected G3d-derived mimotopes allowed the isolation of both stimulatory and inhibitory mimotopes showing differential therapeutic vaccine efficacy. Thus, the mimotope library is a promising vaccine immunogen and a reliable source for isolating molecular cancer vaccine components.

The framing of time-dependent machine learning models improves risk estimation among young individuals with acute coronary syndromes.

Acute coronary syndrome (ACS) is a common cause of death in individuals older than 55 years. Although younger individuals are less frequently seen with ACS, this clinical event has increasing incidence trends, shows high recurrence rates and triggers considerable economic burden. Young individuals with ACS (yACS) are usually underrepresented and show idiosyncratic epidemiologic features compared to older subjects. These differences may justify why available risk prediction models usually penalize yACS with higher false positive rates compared to older subjects. We hypothesized that exploring temporal framing structures such as prediction time, observation windows and subgroup-specific prediction, could improve time-dependent prediction metrics. Among individuals who have experienced ACS (nglobal_cohort = 6341 and nyACS = 2242), the predictive accuracy for adverse clinical events was optimized by using specific rules for yACS and splitting short-term and long-term prediction windows, leading to the detection of 80% of events, compared to 69% by using a rule designed for the global cohort.

Rationale and design of the Brazilian diabetes study: a prospective cohort of type 2 diabetes.

BACKGROUND: Optimal control of traditional risk factors only partially attenuates the exceeding cardiovascular mortality of individuals with diabetes. Employment of machine learning (ML) techniques aimed at the identification of novel features of risk prediction is a compelling target to tackle residual cardiovascular risk. The objective of this study is to identify clinical phenotypes of T2D which are more prone to developing cardiovascular disease. METHODS: The Brazilian Diabetes Study is a single-center, ongoing, prospective registry of T2D individuals. Eligible patients are 30 years old or older, with a confirmed T2D diagnosis. After an initial visit for the signature of the informed consent form and medical history registration, all volunteers undergo biochemical analysis, echocardiography, carotid ultrasound, ophthalmologist visit, dual x-ray absorptiometry, coronary artery calcium score, polyneuropathy assessment, advanced glycation end-products reader, and ambulatory blood pressure monitoring. A 5-year follow-up will be conducted by yearly phone interviews for endpoints disclosure. The primary endpoint is the difference between ML-based clinical phenotypes in the incidence of a composite of death, myocardial infarction, revascularization, and stroke. Since June/2016, 1030 patients (mean age: 57 years, diabetes duration of 9.7 years, 58% male) were enrolled in our study. The mean follow-up time was 3.7 years in October/2021. CONCLUSION: The BDS will be the first large population-based cohort dedicated to the identification of clinical phenotypes of T2D at higher risk of cardiovascular events. Data derived from this study will provide valuable information on risk estimation and prevention of cardiovascular and other diabetes-related events. CLINICALTRIALS.GOV IDENTIFIER: NCT04949152.

Education and armed conflict: consequences on the population of children and adolescents in Colombia: editorial / Educación y conflicto armado: consecuencias en la población de niños, niñas y adolescentes en Colombia: editorial

Throughout history, the right to education has been a determining and unquestionably necessary factor for the improvement of people; however, as a socio-political process, it has not been free of adversities. In this sense, for more than six decades, Colombia has lived in the context of an internal armed conflict

A lo largo de la historia, el derecho a la educación ha sido un factor determinante e indiscutiblemente necesario para la superación de las personas; sin embargo, como proceso sociopolítico, no ha estado exento de adversidades. En este sentido, desde hace más de seis décadas, Colombia vive en el contexto de un conflicto armado interno

Machine Learning Improves the Identification of Individuals With Higher Morbidity and Avoidable Health Costs After Acute Coronary Syndromes.

OBJECTIVES: Traditional risk scores improved the definition of the initial therapeutic strategy in acute coronary syndrome (ACS), but they were not designed for predicting long-term individual risks and costs. In parallel, attempts to directly predict costs from clinical variables in ACS had limited success. Thus, novel approaches to predict cardiovascular risk and health expenditure are urgently needed. Our objectives were to predict the risk of major/minor adverse cardiovascular events (MACE) and estimate assistance-related costs. METHODS: We used a 2-step approach that: (1) predicted outcomes with a common pathophysiological substrate (MACE) by using machine learning (ML) or logistic regression (LR) and compared with existing risk scores; (2) derived costs associated with noncardiovascular deaths, dialysis, ambulatory-care-sensitive-hospitalizations (ACSH), strokes, and MACE. With consecutive ACS individuals (n = 1089) from 2 cohorts, we trained in 80% of the population and tested in 20% using a 4-fold cross-validation framework. The 29-variable model included socioeconomic, clinical/lab, and coronarography variables. Individual costs were estimated based on cause-specific hospitalization from the Brazilian Health Ministry perspective. RESULTS: After up to 12 years follow-up (mean = 3.3 ± 3.1; MACE = 169), the gradient-boosting machine model was superior to LR and reached an area under the curve (AUROC) of 0.891 [95% CI 0.846-0.921] (test set), outperforming the Syntax Score II (AUROC = 0.635 [95% CI 0.569-0.699]). Individuals classified as high risk (>90th percentile) presented increased HbA1c and LDL-C both at <24 hours post-ACS and 1-year follow-up. High-risk individuals required 33.5% of total costs and showed 4.96-fold (95% CI 3.71-5.48, P < .00001) greater per capita costs compared with low-risk individuals, mostly owing to avoidable costs (ACSH). This 2-step approach was more successful for finding individuals incurring high costs than predicting costs directly from clinical variables. CONCLUSION: ML methods predicted long-term risks and avoidable costs after ACS.

A differential sex-specific pattern of IgG2 and IgG4 subclasses of anti-drug antibodies (ADAs) induced by glatiramer acetate in relapsing-remitting multiple sclerosis patients.

BACKGROUND: Glatiramer acetate (GA) is a drug for Multiple Sclerosis (MS) treatment. However, its administration induces anti-drug antibodies (ADA). This research evaluated the sex differences in humoral response against GA in RR-MS patients METHODS: We analyzed 69 RR-MS patients, 43 treated with GA and 26 treated with IFN-ß. In all cases, the serum concentration of IgG antibodies was determined by UPLC, whereas the levels of IgG subclasses (1-4) of anti-GA antibodies and the concentration of IL-6 were detected by Multiplex and IL-10, and IFN-γ were detected by ELISA. RESULTS: The total concentration of IgG antibodies in patients did not differ between treatments, whereas the IgG levels of ADA were higher in male and female patients treated with GA (P ≤ 0.0001). The subclasses of IgG anti-GA antibodies were as follows: IgG4>>IgG3>IgG1>IgG2. Statistical analysis showed differences in the IgG2 (P ≤ 0.01) and IgG4 (P ≤ 0.0001) subclasses by sex in RR-MS patients. Levels of IgG1 subclass in male patients correlated positively with the circulatory levels of IL-6 (rs = 0.587, P ≤ 0.04) and IFN-γ (rs = 0.721, P ≤ 0.001), while IgG2 subclass levels in female patients correlated with serum levels of IFN-γ (rs = 0.628, P ≤ 0.0006). Statistical analysis did not detect correlations between the levels of IgG (1-4) subclasses of anti-GA antibodies and the evaluated clinical parameters. CONCLUSION: This study showed differences in the levels of IgG2 and IgG4 subclasses of ADA between male and female RR-MS patients. Further studies are necessary to take advantage of the clinical potential of this finding.

Development of Functional Antibodies Directed to Human Dialyzable Leukocyte Extract (Transferon®).

Transferon® is an immunomodulator made of a complex mixture of peptides from human dialyzable leucocyte extracts (hDLEs). Development of surrogate antibodies directed to hDLE is an indispensable tool for studies during process control and preclinical trials. These antibodies are fundamental for different analytical approaches, such as identity test and drug quantitation, as well as to characterize its pharmacokinetic and mechanisms of action. A previous murine study showed the inability of the peptides of Transferon® to induce antibody production by themselves; therefore, in this work, two approaches were tested to increase its immunogenicity: chemical conjugation of the peptides of Transferon® to carrier proteins and the use of a rabbit model. Bioconjugates were generated with Keyhole Limpet Hemocyanin (KLH) or Bovine Serum Albumin (BSA) through maleimide-activated carrier proteins. BALB/c mice and New Zealand rabbits were immunized with Transferon® conjugated to KLH or nonconjugated Transferon®. Animals that were immunized with conjugated Transferon® showed significant production of antibodies as evinced by the recognition of Transferon®-BSA conjugate in ELISA assays. Moreover, rabbits showed higher antibody titers when compared with mice. Neither mouse nor rabbits developed antibodies when immunized with nonconjugated Transferon®. Interestingly, rabbit antibodies were able to partially block IL-2 production in Jurkat cells after costimulation with Transferon®. In conclusion, it is feasible to elicit specific and functional antibodies anti-hDLE with different potential uses during the life cycle of the product.

A data-driven approach to referable diabetic retinopathy detection.

Prior art on automated screening of diabetic retinopathy and direct referral decision shows promising performance; yet most methods build upon complex hand-crafted features whose performance often fails to generalize. OBJECTIVE: We investigate data-driven approaches that extract powerful abstract representations directly from retinal images to provide a reliable referable diabetic retinopathy detector. METHODS: We gradually build the solution based on convolutional neural networks, adding data augmentation, multi-resolution training, robust feature-extraction augmentation, and a patient-basis analysis, testing the effectiveness of each improvement. RESULTS: The proposed method achieved an area under the ROC curve of 98.2% (95% CI: 97.4-98.9%) under a strict cross-dataset protocol designed to test the ability to generalize - training on the Kaggle competition dataset and testing using the Messidor-2 dataset. With a 5 × 2-fold cross-validation protocol, similar results are achieved for Messidor-2 and DR2 datasets, reducing the classification error by over 44% when compared to most published studies in existing literature. CONCLUSION: Additional boost strategies can improve performance substantially, but it is important to evaluate whether the additional (computation- and implementation-) complexity of each improvement is worth its benefits. We also corroborate that novel families of data-driven methods are the state of the art for diabetic retinopathy screening. SIGNIFICANCE: By learning powerful discriminative patterns directly from available training retinal images, it is possible to perform referral diagnostics without detecting individual lesions.

Diseño y validación de un simulador híbrido de paracentesis / Design and validation of paracentesis hybrid simulator

La enseñanza médica por simulación es un método eficaz para el aprendizaje; mejora la adquisición de competencias, la práctica repetitiva y elimina riesgos para el paciente. Los simuladores de paracentesis existentes, tienen un alto costo y bajo nivel de realismo, por lo que se diseñó y elaboró un simulador híbrido de bajo costo. Se realizó un video representativo de un escenario clínico apropiado, que demuestra la técnica completa de paracentesis. Veinte médicos especialistas observaron el video y realizaron el procedimiento en simulador, posteriormente se aplicó un cuestionario sobre el realismo del simulador, utilidad del video e importancia del procedimiento en la educación médica. Los resultados mostraron que el 85% de los participantes considera que el simulador favoreció la experiencia de aprendizaje. El 90% consideró la apariencia clínica muy adecuada. El 100% opina que la adquisición de la habilidad de paracentesis es relevante en alumnos de pregrado

Medical education by simulation is an effective method for learning; It improves competence acquisition, repetitive practice and eliminates risks for the patient. Existing paracentesis simulators have a high cost and low level of realism, so a low cost simulator was designed and developed. A representative video of a suitable clinical scenario was demonstrated, demonstrating the complete paracentesis technique. 20 medical specialists observed the video and performed the procedure in the simulator, later applied a questionnaire on the realism of the simulator, the utility of the video, and the importance of the procedure in medical education. The results show that 85% of the participants consider the simulator favored the learning experience. 90% considered the clinical appearance to be adequate. 100% believe that the acquisition of paracentesis ability is relevant in undergraduate students

A Machine Learning Application Based in Random Forest for Integrating Mass Spectrometry-Based Metabolomic Data: A Simple Screening Method for Patients With Zika Virus.

Recent Zika outbreaks in South America, accompanied by unexpectedly severe clinical complications have brought much interest in fast and reliable screening methods for ZIKV (Zika virus) identification. Reverse-transcriptase polymerase chain reaction (RT-PCR) is currently the method of choice to detect ZIKV in biological samples. This approach, nonetheless, demands a considerable amount of time and resources such as kits and reagents that, in endemic areas, may result in a substantial financial burden over affected individuals and health services veering away from RT-PCR analysis. This study presents a powerful combination of high-resolution mass spectrometry and a machine-learning prediction model for data analysis to assess the existence of ZIKV infection across a series of patients that bear similar symptomatic conditions, but not necessarily are infected with the disease. By using mass spectrometric data that are inputted with the developed decision-making algorithm, we were able to provide a set of features that work as a "fingerprint" for this specific pathophysiological condition, even after the acute phase of infection. Since both mass spectrometry and machine learning approaches are well-established and have largely utilized tools within their respective fields, this combination of methods emerges as a distinct alternative for clinical applications, providing a diagnostic screening-faster and more accurate-with improved cost-effectiveness when compared to existing technologies.

Process signatures in glatiramer acetate synthesis: structural and functional relationships.

Glatiramer Acetate (GA) is an immunomodulatory medicine approved for the treatment of multiple sclerosis, whose mechanisms of action are yet to be fully elucidated. GA is comprised of a complex mixture of polypeptides with different amino acid sequences and structures. The lack of sensible information about physicochemical characteristics of GA has contributed to its comprehensiveness complexity. Consequently, an unambiguous determination of distinctive attributes that define GA is of highest relevance towards dissecting its identity. Herein we conducted a study of characteristic GA heterogeneities throughout its manufacturing process (process signatures), revealing a strong impact of critical process parameters (CPPs) on the reactivity of amino acid precursors; reaction initiation and polymerization velocities; and peptide solubility, susceptibility to hydrolysis, and size-exclusion properties. Further, distinctive GA heterogeneities were correlated to defined immunological and toxicological profiles, revealing that GA possesses a unique repertoire of active constituents (epitopes) responsible of its immunological responses, whose modification lead to altered profiles. This novel approach established CPPs influence on intact GA peptide mixture, whose physicochemical identity cannot longer rely on reduced properties (based on complete or partial GA degradation), providing advanced knowledge on GA structural and functional relationships to ensure a consistent manufacturing of safe and effective products.

Transferon™, a peptide mixture with immunomodulatory properties is not immunogenic when administered with various adjuvants.

Transferon, a human dialyzable leukocyte extract (hDLE), is a biotherapeutic that comprises a complex mixture of low-molecular-weight peptides (< 10 kDa) and is used to treat diseases with an inflammatory component. Some biotherapeutics, including those composed of peptides, can induce anti-drug antibodies (ADA) that block or diminish their therapeutic effect. Nevertheless, few studies have evaluated peptide-derived drug immunogenicity. In this study, the immunogenicity of Transferon was examined in a murine model during an immunization scheme using the following adjuvants: Al(OH)3, incomplete Freund's adjuvant (IFA), or Titermax Gold. The inoculation scheme entailed three routes of administration (intraperitoneal, Day 1; subcutaneous, Day 7; and intramuscular, Day 14) using 200 µg Transferon/inoculation. Serum samples were collected on Day 21. Total IgG levels were quantitated by affinity chromatography, and specific antibodies against components of Transferon were analyzed by dot-blot and ELISA. Ovalbumin (OVA, 44 kDa) and peptides from hydrolyzed collagen (PFHC, < 17 kDa) were used as positive and negative controls, respectively, in the same inoculation scheme and analyses for Transferon. OVA, PFHC, and Transferon increased total IgG concentrations in mice. However, only IgG antibodies against OVA were detected. Based on the results, it is concluded that Transferon does not induce generation of specific antibodies against its components in this model, regardless of adjuvant and route of administration. These results support the safety of Transferon by confirming its inability to induce ADA in this animal model.

Beyond Lesion-Based Diabetic Retinopathy: A Direct Approach for Referral.

Diabetic retinopathy (DR) is the leading cause of blindness in adults, but can be managed if detected early. Automated DR screening helps by indicating which patients should be referred to the doctor. However, current techniques of automated screening still depend too much on the detection of individual lesions. In this study, we bypass lesion detection, and directly train a classifier for DR referral. Additional novelties are the use of state-of-the-art mid-level features for the retinal images: BossaNova and Fisher Vector. Those features extend the classical Bags of Visual Words and greatly improve the accuracy of complex classification tasks. The proposed technique for direct referral is promising, achieving an area under the curve of 96.4%, thus, reducing the classification error by almost 40% over the current state of the art, held by lesion-based techniques.

Pornography classification: The hidden clues in video space-time.

As web technologies and social networks become part of the general public's life, the problem of automatically detecting pornography is into every parent's mind - nobody feels completely safe when their children go online. In this paper, we focus on video-pornography classification, a hard problem in which traditional methods often employ still-image techniques - labeling frames individually prior to a global decision. Frame-based approaches, however, ignore significant cogent information brought by motion. Here, we introduce a space-temporal interest point detector and descriptor called Temporal Robust Features (TRoF). TRoF was custom-tailored for efficient (low processing time and memory footprint) and effective (high classification accuracy and low false negative rate) motion description, particularly suited to the task at hand. We aggregate local information extracted by TRoF into a mid-level representation using Fisher Vectors, the state-of-the-art model of Bags of Visual Words (BoVW). We evaluate our original strategy, contrasting it both to commercial pornography detection solutions, and to BoVW solutions based upon other space-temporal features from the scientific literature. The performance is assessed using the Pornography-2k dataset, a new challenging pornographic benchmark, comprising 2000 web videos and 140h of video footage. The dataset is also a contribution of this work and is very assorted, including both professional and amateur content, and it depicts several genres of pornography, from cartoon to live action, with diverse behavior and ethnicity. The best approach, based on a dense application of TRoF, yields a classification error reduction of almost 79% when compared to the best commercial classifier. A sparse description relying on TRoF detector is also noteworthy, for yielding a classification error reduction of over 69%, with 19× less memory footprint than the dense solution, and yet can also be implemented to meet real-time requirements.

A pilot study of nimotuzumab plus single agent chemotherapy as second- or third-line treatment or more in patients with recurrent, persistent or metastatic cervical cancer.

Nimotuzumab is a humanized IgG1 monoclonal antibody against the EGFR extracellular domain that has been evaluated in solid tumors as a single agent or in combination with chemotherapy and radiation. Cervical cancer patients who are refractory or progressive to first-line chemotherapy have a dismal prognosis, and no second- or third-line chemotherapy is considered standard. This pilot trial aimed to evaluate the efficacy and safety of nimotuzumab in 17 patients with pre-treated advanced refractory or progressive cervical cancer. Nimotuzumab was administered weekly at 200 mg/m(2) as single agent for 4 weeks (induction phase), then concurrent with 6 21-day cycles of gemcitabine (800 mg/m(2)) or cisplatin (50 mg/m(2)) for 18 weeks (concurrent phase) and then once every 2 weeks (maintenance phase). Nimotuzumab could be continued beyond disease progression. Seventeen patients were accrued and evaluated for safety and efficacy. The median number of nimotuzumab applications was 20 (5-96). The median number of chemotherapy cycles administered was 6 (1-6). No toxicity occurred during induction and maintenance phases (single agent nimotuzumab). In the concurrent phase, grade 3 toxicity events observed were leucopenia, anemia and diarrhea in 11.7%, 5.8% and 11.7% respectively. No complete or partial responses were observed. The stable disease (SD) rate was 35%. The median PFS and OS rates were 163 days (95% CI, 104 to 222), and 299 days (95% IC, 177 to 421) respectively. Nimotuzumab is well tolerated and may have a role in the treatment of advanced cervical cancer.

Automatic Diabetic Retinopathy detection using BossaNova representation.

The biomedical community has shown a continued interest in automated detection of Diabetic Retinopathy (DR), with new imaging techniques, evolving diagnostic criteria, and advancing computing methods. Existing state of the art for detecting DR-related lesions tends to emphasize different, specific approaches for each type of lesion. However, recent research has aimed at general frameworks adaptable for large classes of lesions. In this paper, we follow this latter trend by exploring a very flexible framework, based upon two-tiered feature extraction (low-level and mid-level) from images and Support Vector Machines. The main contribution of this work is the evaluation of BossaNova, a recent and powerful mid-level image characterization technique, which we contrast with previous art based upon classical Bag of Visual Words (BoVW). The new technique using BossaNova achieves a detection performance (measured by area under the curve - AUC) of 96.4% for hard exudates, and 93.5% for red lesions using a cross-dataset training/testing protocol.